MIT flight planner avoids moving obstacles in 12 UAV tests

MIT researchers developed a flight planner that avoided all dynamic obstacles in 12 real UAV test flights, using onboard computing and sensors to replan trajectories. Called SANDO, the system also reached its goal faster than several competing systems in simulations while avoiding collisions in all tested environments. The planner builds a flight corridor around regions that moving obstacles could reach over time, then optimizes the route within it. Researchers mathematically proved collision avoidance for moving obstacles in unknown environments. The guarantee requires knowing the maximum speed obstacles could reach; their future direction need not be known.